Draft
Conversation
EMZEDI
reviewed
May 13, 2025
jobs/ccpo/ccpo_ccpo_multi_gpu.sh
Outdated
| --dataset_name $dataset_name \ | ||
| --sft_model_path Qwen/Qwen2-0.5B-Instruct \ | ||
| --value_model_path Shahradmz/Qwen2-0.5B-Instruct_${dataset_name}_REWARD_0 \ | ||
| --reward_model_path Shahradmz/Qwen2-0.5B-Instruct_${dataset_name}_REWARD \ |
Collaborator
There was a problem hiding this comment.
Suggested change
| --reward_model_path Shahradmz/Qwen2-0.5B-Instruct_${dataset_name}_REWARD \ | |
| --reward_model_path Shahradmz/Qwen2-0.5B-Instruct_CPPO_REWARD \ |
Collaborator
|
BUG: Need to pass the tokenizer when doing data loading for DPO, PPO, CPPO, and other benchmarks, otherwise the chat template is not going to be applied to the data... |
reward hacking can be solved by increasing the KL coefficient but it means that the reward model needs to be retrained after increasing the divergence between the chosen and rejected responses
* Add ppo ewc jobs * Update benchmark entry points * Update trainers * Typo in CPPO job * Update jobs output dir
* FIx dpo job reward model * Add dpo ewc jobs * Rename
Jacob-Chmura
commented
Jun 1, 2025
| ContinualPPOTrainer.ds_wrapped_models = self.model | ||
| else: | ||
| self.ds_wrapped_models = self.model | ||
| elif False: |
Member
Author
There was a problem hiding this comment.
I can go ahead and drop these conditionals throughout this PR to simplify all our benchmarking code
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.